excavations at former milldown school ......introduction in march and april 2015, cotswold...

EXCAVATIONS AT FORMER MILLDOWN SCHOOL, BLANDFORD FORUM, DORSET, 2015

JOE WHELAN and RICHARD MASSEY

Offprint from the

Proceedings of the Dorset Natural History & Archaeological Society

Volume 137

2016

Hon. Editor, Dr Paul Lashmar

Introduction

In March and April 2015, Cotswold Archaeology (CA) carried out an archaeological excavation on the behalf of Bellway Homes Ltd, at the former Milldown School, Blandford Forum, Dorset (centred on NGR: ST 88303 07385; Fig. 1) (Cotswold Archaeology 2015).

The development site was 1.2 ha in extent, and situated within the grounds of the former Milldown

School, Blandford Forum, Dorset (Fig. 1). The site slopes gently to the south and south-west, at an elevation averaging 58 m above Ordnance Datum. The underlying geology comprises clay-with-flints overlying Upper Chalk.

Prior to evaluation in 2012 (AC 2012), no archaeological finds or features had been recorded within the site, although an Early Bronze

Excavations at former Milldown School, Blandford Forum, Dorset, 2015JOE WHELAN and RICHARD MASSEY

with contributions from E.R. McSloy, Jacky Sommerville, Sarah Cobain and Andy Clarke

Abstract

An archaeological excavation was undertaken by Cotswold Archaeology in March and April 2015, at the Former Milldown School, Blandford Forum, Dorset. The fieldwork consisted of five separate excavation areas (Areas A – E), which targeted a series of features identified in two previous evaluations. Excavation Area D identified two sides of a Middle Bronze Age ditched enclosure, with evidence for relatively short-lived domestic settlement. Finds, including moderate quantities of pottery, confirmed the domestic character of the enclosure. The relatively well-preserved and decorated pottery group was representative of the local Deverel-Rimbury tradition, dating to c.1600 – 1000 BC, and represents a significant addition to the regional record. Limited plant macrofossil evidence recovered from the site indicated that the processing of cereals took place within the enclosure, and faunal remains included evidence for the rearing of cattle and sheep or goats. Tentatively-dated later prehistoric pottery from Area D, and early Roman pottery from Area A, indicated later episodes of activity on the site.

Figure 1. Site location

219

Age unenclosed settlement has been recorded approximately 400 m to the south-west, on the site of the current Milldown School (AC 2008).

The 2012 evaluation comprised eight trenches, and recorded an enclosure ditch (F506), which conformed to the ditches, 177 and 178, recorded by the 2015 investigation. Further evaluation of the site in November 2014 (CA 2014) confirmed a Bronze Age date, which contradicted the late Neolithic date suggested by the 2012 evaluation (AC 2012).

Five individual excavation areas were targeted on the results of the two previous archaeological evaluations (AC 2012 & CA 2014). Area A measured approximately 10 x 12 m, Area B measured c. 12 x 15 m, Area C measured c. 15 x 22 m, Area D measured c. 15 x 45 m and Area E measured c. 10 x 15 m (Fig 2). Exposed archaeological features were hand-excavated and planned, following the machine stripping of topsoil within individual excavation areas. Nine environmental samples were retained from seven features. This paper summarises a report (CA 2015) which can be found on the Cotswold Archaeology website: http://www.cotswoldarchaeology.co.uk/. It is intended that the archive will be deposited with Dorset County Museum.

The Excavation (Figs 2-3)Drift geology varied across the site, with underlying chalk exposed further downslope beneath periglacial sands and gravels, and clay-with-flints. Natural features were recorded in all five excavation areas. In addition to probable tree-throw hollows, a small number of natural hollows filled with residual subsoil were recorded in Areas A, C and D. A deeper horizon of colluvium, of up to a metre in depth, was recorded in the south of Area E, on the slope of a shallow combe.

Area D: Earlier Prehistoric

A small number of residual flint bladelets, blades and possible cores were recovered, principally from pit 158 and ditch 178. These items, particularly the bladelets, were typical of Mesolithic or early Neolithic flint-working.

Area D: Early Bronze Age

Seven sherds of Early Bronze Age pottery, including four of Beaker fabric, were recovered from F211 of Trench 2, of the 2012 evaluation (AC 2012, 3-4).

Feature 211 represents the southernmost excavated extent of the Bronze Age ditch 178. Beaker sherds in this context may perhaps represent residual material from disturbed earlier burials.

Area E: Bronze Age

Area E, located on a slope, targeted CA Evaluation Trench 1. It contained a small pit, 108, containing worked flint, beneath a deep colluvial deposit. The excavation of Area E revealed a further small, undated, oval pit, 119. Re-stripping to remove remaining colluvium confirmed that no further features underlay this deposit.

Area D Middle Bronze Age (Fig. 2)

Area D targeted an area which had previously been evaluated by six separate trenches during the 2012 and 2014 evaluations. These evaluations had identified the south-east and north-east sides (ditches 177 and 178) of a ditched enclosure of probable rectilinear plan. The enclosure contained an internal alignment of three small pits or post holes (132, 149 and 134), and a single large pit (161) of Middle Bronze Age date. Two large quarry pits (158, 130) were also recorded outside the enclosure in Area D.

The excavated lengths of ditches 177 and 178 were 8.2 m and 27.3 m respectively. The U-profile ditches averaged 0.55m in width, and 0.35m in depth, with an entranceway, 1.9 m wide, on the north-east side. All fills comprised a uniform, yellow-brown clay-silt, with common flints.

Terminal 153 and Slot 156 were hand-excavated within ditch 177, both of which contained single fills. Terminal 153 contained Bronze Age pottery, and two circular clay loom weights. A quantity of burnt flint, totalling 3.8 kg, was recovered from slot 156, along with non-local stone, items of worked flint and Bronze Age pottery.

Terminal 151 and Slots 165 and 172 were hand-excavated within ditch 178. Fill 166, of Slot 165, produced a copper-alloy awl and Bronze Age pottery and burnt flint. Slot 174 was originally excavated during the 2012 evaluation, producing 166 sherds of pottery weighing 854 g. Although originally identified as grooved ware of later Neolithic date (AC 2012, 11), Slot 174 was extended to recover a further sample of this material. This was subsequently identified as Fabric Type GR1, of a single vessel and of Middle Bronze Age date, and therefore broadly

Joe Whelan and Richard Massey

220

Figure 2. Plan of site with all features

Excavations at former Milldown School, Blandford Forum, Dorset, 2015

221

contemporary with other pottery recorded from ditch 178.

An alignment of three small, U-profile circular pits, or large postholes, extended westwards from the entranceway on the north-east side into the interior of the enclosure. Pits 132, 134 and 149 each measured between 0.60 – 0.65 m in diameter, and

between 0.19 – 0.26 m in depth. The fills of pits 132 and 149 contained Middle Bronze Age pottery, with worked flint in the latter.

Pit 161 comprised a large, circular, former storage pit, which had been cut into the chalk bedrock within the enclosure (Fig. 3). The pit measured 1.99 m by 1.84 m in plan, with a depth of 1.61 m. Its upper

Figure 3. Sections and photographs

222


The Finds

Worked Flint (Jacky Sommerville)

A total of 196 worked flints (2.747 kg), and 240 pieces of burnt, unworked flint (10.9 kg) was recovered from the excavation of twenty-three separate deposits.

Cortex was in an abraded condition on 25% of items, indicating limited use of secondary sources, such as river or beach pebbles. Eight items displayed recortication on dorsal faces or butts, indicating reuse of earlier tools. The great majority (95%) of items were made on flint of moderate quality, most of which displayed coarse, cherty inclusions.

Of the worked flints recovered from deposits of Middle Bronze Age, and later, date, 71% were minimally edge-damaged, and 89% minimally rolled, suggesting that a large proportion of these items were recovered stratified.

Flint debitage (flakes, blades, bladelets and chips) totalled 165 items. Ten flakes displayed evidence of utilisation along one or more edges, and 22% of flakes (some retouched) ended in a hinge termination, suggesting unskilled knapping technique (Whittaker 1994, 109) typical of Bronze Age assemblages (Ford et al. 1984, 163). Only two debitage flakes (1%) were primary, 110 (75%) secondary, and 34 (23%) tertiary.

Eight cores were recovered, comprising three dual-platform and five multi-platform types. The majority had been used to produce flakes, although two also featured possible blade scars, both residual items from the Roman-dated fill, 103, of ditch 102. Retouched flints totalled twenty-three which, at 12%, was a relatively high proportion. The majority comprised retouched flakes, notched flakes and blades, and spurred pieces. Scrapers totalled five, including a combination end-and-side scraper/notched flake, from the topsoil, and end and side scrapers made from broken flakes. A knife, from fill 114, of undated gully 113, featured partial bifacial retouch.

The largely stratified assemblage of worked flint was diagnostic of Bronze Age date, with a notable predominance of processing tools suggesting domestic use. A residual component (26%) was largely undiagnostic, a proportion of which is likely to have derived from Middle Bronze Age activity on site. A small number of items (bladelets, blades and possible blade cores) suggested activity on the site during the Mesolithic or Early Neolithic periods.

profile displayed undercutting sides. A single, deep post hole (170) was recorded in an off-centre position within the base of this pit. The secondary fill, 163, of Pit 161, representing a third of the total pit volume, was a rich humic deposit, possibly representing cess material. The remaining fills comprised a mixture of chalk weathering deposits (162), and a more uniform fill suggestive of rapid backfilling (164). The fills contained Middle Bronze Age pottery, worked flint and animal bone.

Area D Later prehistoric periods Figs 2 and 3

Two large, oval pits were recorded outside Enclosure 155. Pit 130 comprised an oval, flat-bottomed pit, of up to 4 m in length. Irregular pit 158 was located immediately to the south-east of enclosure ditch 178, and displayed a large number of tapered stake holes within its chalk base.

Pit 158 contained two fills, the primary fill, 159, comprised a dark, grey-brown silt, and the secondary fill, 160, contained eight sherds of a distinctive thinner-walled, flint-tempered fabric, which may represent a transient later Bronze Age or early Iron Age phase of activity on the site. Pit 158 also produced thirty-two items of worked flint.

Area A: Roman

Area A targeted Evaluation Trench 9, which contained a single undated pit (179). Ditch or gully feature 102 extended beyond the Area A trench to the north-west, and measured approximately 5 m by 1.7 m in plan, with a maximum depth of 0.16 m (Fig. 3). It contained a single fill, 103, which contained a small quantity of Roman pottery of first- to second-century AD date, together with burnt flint. Ditch 113 extended beyond Area A, to the north, and measured 2 m in length by 0.70 m in width, with a maximum depth of 0.32 m. Its fill, 114, contained a flint knife.

Area C: Medieval/post-medieval

Area C targeted evaluation Trenches 5, 6 and 8, within which two small, undated pits and several tree-throw hollows were recorded. The excavation of Area C revealed a further five natural hollows, in addition to a post-medieval gully, 143, in the south-east corner of the trench.

223


Figure 4. Illustrated pottery vessels

224


The Pottery (E.R. McSloy)

Pottery amounting to 500 sherds (5647 g) was recovered. The majority was hand-recovered by excavation, with thirty sherds retrieved from soil sample residues.

The great majority of the assemblage was recorded from features in Area D. Almost all this material, and further small quantities from Area C, belongs to the Middle Bronze Age Deverel-Rimbury tradition, dating to c. 1600 – 1000 BC. The pottery has been fully recorded and quantified by sherd count and weight per fabric, and vessel form, rim morphology, type and location of decoration when present. Use/wear evidence and sherd thickness were also recorded. Pottery fabrics (described in summary below) are defined on the basis of primary inclusion type and inclusion coarseness/abundance.

Prehistoric

Table 1 shows the quantities of material relating to Area C/D features. Its condition is typically good, with minimum surface loss/abrasion noted and including a number of vessels reconstructable to full profile, or represented by multiple joining sherds. The largest and best-preserved pottery groups are those from pit 161 and ditch terminal 154, which together make up 72 % of the assemblage by sherd count.

A small group of pottery (eight sherds, weighing 36 g), from Area D pit 158, was different in character to the bulk of the assemblage, and is described ‘late prehistoric’. A thin-walled vessel in sandy fabric QF1 was amongst two rim sherds (P25–P26) from this group. The date of this material is unclear, although a later Bronze Age or earlier Iron Age date is thought most likely.

Fabrics: Middle Bronze Age

F1 ‘Standard’ medium-coarse, flint-tempered. Common, moderately-sorted flint inclusions (1–3 mm); some iron oxide.

F2 Medium-coarse flint, with grog. Common to sparse, moderately-sorted flint (1–2 mm); sparse grog (1–2 mm) and sparse grog.

F3 Coarser flint-tempered. Common, poor-sorted flint (2–5 mm).

F4 Fine flint-tempered. Common or abundant, well-sorted fine flint (

straight profiles, although this is not always clearly evident from smaller sherds. The rims among such vessels exhibit limited morphological variation, most being simple rounded or squared, and occasionally expanded (T-shaped or internally thickened). All but three of the barrel/bucket ‘urns’ (P9, P11 and P14) exhibit decoration, or feature lugs. Most commonly, decoration is in the form of single or double rows of fingernail impressions (P4, P7, P8, P10, P15, P17–P19), which are sometimes coupled with fingernail impressions, or slashes to the rim tops (P7 and P15). A small number of vessels exhibit greater elaboration in the form of fingernail-impressed ‘arcs’ below the rim (P7) and, most unusually, as columns of impressions at the neck, executed with square and rectangular implements (P3). Applied or ‘pinched-up’ features occur as simple rounded or ’lenticular’ lugs (P5, P12, P20), and plain or finger-impressed horizontal strips (P1, P3, P24).

Globular ‘urns’ make up a minority of the identifiable vessels (5), which include examples from pit 161 and ditch 174, where full vessel profiles are preserved. Such forms are a characteristic element among Deverel Rimbury assemblages, and are typically smaller and with bulbous ‘belly’ and constricted neck zone. All of the identified globular urn vessels feature decoration in the form of deeply-incised horizontal bands, which, along with finer fabrics, is a feature of ‘Type 2’ Globular urns within the south Dorset area (Calkin 1964, 24). Vessel P2 exhibits additional scored decoration below the neck, possibly in the form of a wavy line. The two more complete vessels, P6 and P13, both feature two or more applied lugs, which are perforated (horizontally) only on P13.

Form and decoration: Middle Bronze Age

The assemblage included rim sherds from a minimum twenty vessels and these, together with selected other featured (decorated body sherds and base sherds), are described in the catalogue. The majority of vessels comprise neckless forms historically referred as ‘Urn’ classes, reflecting their widespread use as cinerary containers. Such vessels are typically large, mostly with rim diameters in the range 220–280mm. A distinction is made (Table 2) between barrel and bucket ‘Urns’ based on curving and

is significantly greater among the more coarsely (flint) gritted types F1 and F3, with the majority (83%) in excess of 9 mm, and to a maximum of 18 mm; compared to 63% for the other types combined, and to a maximum of 15 mm.

Chart 1. The Relative Proportions of Middle Bronze Age pottery fabrics, by percentage

Table 2. Prehistoric pottery summary by feature . Quantities as sherd count and weight (grams)Feature> 132 149 151 153 156 158 161 165 172 174 Totals

fabric Ct Wt Ct Wt Ct Wt Ct Wt Ct Wt Ct Wt Ct Wt Ct Wt Ct Wt Ct Wt Ct Wt Ct Wt

F1 20 135 2 10 109 481 9 17 81 2376 1 1 1 3 223 3023F2 37 153 37 153F3 7 96 28 411 1 3 7 166 43 676F4 1 37 73 544 3 12 77 593F5 12 81 2 10 1 8 15 99GR1 1 1 19 510 6 102 43 204 69 817GRF 5 59 2 11 7 70

QF1 4 18 4 18

F6 4 18 4 18

Total 1 1 20 135 2 10 45 286 246 2086 12 30 8 36 100 2675 1 1 1 3 43 204 479 5467

226


Evidence for use

Incidence of carbonaceous or other residues is sparse in the assemblage, and was recorded on only nine sherds with external ‘sooting’, and five sherds with internal ‘burnt food’-type deposits. Most sherds preserving residues are not identifiable to vessel form, although, notably among those with sooting is Globular urn P13. Such low incidence is unlikely to be the result of poor preservation (surface survival was good), and may reflect cooking regimes not requiring contact with, or close proximity to, open fire, or conversely the use of pottery primarily for storage or other non-cooking related activities.

One vessel (P9) exhibits a post-firing perforation below its rim, an adaptation which is fairly commonly seen in pottery of Bronze Age date, and representing either a repair or a means of suspension. Vessel P21 is notable in the treatment of its base, where well-crushed flint is embedded in the underside. This feature, which is recorded elsewhere from Late Bronze Age pottery (post-Deverel Rimbury plainwares) from the Thames Valley (Morris forthcoming), may relate to use, as a means of providing increased resilience to hot surfaces .

DiscussionThe larger part of the assemblage is clearly characteristic of the Middle Bronze Age Deverel-Rimbury tradition, a style which defines pottery from south-central England between the sixteenth to twelfth/eleventh centuries BC. There are stylistic affinities, evident among the (Type 2) Globular urns, with a number of Middle Bronze Age groups recorded around the Poole/ Weymouth/Blandford area of South Dorset (Calkin 1964, 23, Fig. 9).

Refinement within the suggested date range is not possible in the absence of a more complex site stratigraphy, and without absolute dating. The elaborate decoration seen with some larger vessels (P3) recalls earlier Urn styles, and may suggest the group dates to the earlier part of the given range. Overall, the similarities in fabric form/decoration across the larger context/feature groups suggests broad contemporaneity. The small quantities of pottery from irregular pit 158 suggest some limited later prehistoric activity on the site. No pottery of confirmed Late Bronze Age date was identified, although the flint-gritted base of vessel P21 (from ditch 153) is representative of a practice which hitherto appears to have only been

recorded from this (later) period, which may represent a transient, post-Deverel-Rimbury phase of c. 1200 – 1000 BC.

The relative abundance of pottery, and the (albeit limited) evidence for its use, suggests a domestic group. The low density of features within the area sampled, and the apparent absence of structures, may suggest a short-lived site or high levels of truncation. Within the Middle Bronze Age assemblage there are indications of the ‘structured deposition’ of large/well-preserved pottery groups which is comparable with evidence elsewhere in assemblages of this period (Woodward 2009, 264). This is most convincingly apparent in ditch terminal group 153, and Pit 161. Together, these groups account for the greater proportion of the pottery recovered from the site, and include larger sherds and vessels re-constructable to full profile (P6 and P13).

Descriptive catalogue (vessels in bold are illustrated)

Pit 161

P2: Fabric F3. Globular urn. Upright/simple rim. Band of horizontal grooves at the neck and curving motif below. Diam. 190 mm; 0.07 EVEs. (fill 164).

P3: Fabric F1. 3 x joining sherds from large vessel (?barrel urn) with pronounced horizontal cordon with fingertip impressions and design to the neck composed of sub-square and horizontal impressions. (fill 164).

P4: Fabric F1. 7 x joining sherds from large, curving profile (barrel urn) and expanded/T-shaped rim. Two lines of horizontal fingernail impressions below rim. 0.10 EVEs. (fill 164).

P5: Fabric F3. Large vessel with slightly-curving profile (barrel urn) and simple/rounded rim. Single applied imperforate lug below rim. Diam. 180 mm; 0.21 EVEs. (fill 164).

P7: Fabric F1. Mid-sized vessel with slightly-curving profile (barrel urn) and expanded/T-shaped rim. Oblique-scored decoration to rim top; row of fingernail impressions below rim and arc of fingernail impressions under this. Diam. 260 mm; 0.24 EVEs. (fill 163).

P8: Fabric F1. Large vessel with straight profile (bucket urn) and simple/squared rim. Row of fingernail impressions below rim. Diam. 280mm; 0.10 EVEs. (fill 163).

227


P9: Fabric F1. Large vessel with straight profile (bucket urn) and simple/rounded rim. Undecorated, but with post-firing perforation below rim. 0.06 EVEs. (fill 163).

P10: Fabric F1. Mid-sized vessel with concave neck (?barrel urn) and externally-expanded rim. Row of fingernail impressions below rim. Diam. 140mm; 0.12 EVEs. (fill 163). Not illustrated.

P11: Fabric F1. Large vessel with curving profile (barrel urn) and expanded/T-shaped rim. Undecorated. Diam. 220 mm; 0.10 EVEs. (fill 163). Not illustrated.

P12: Fabric F1. Mid-sized vessel with curving profile (barrel urn) and simple/rounded rim. Single low imperforate lug below rim. Diam. 180 mm; 0.12 EVEs. (fill 163).

Ditch 153

P13: Fabric GRF. Globular urn (full profile). Upright/simple rim, rounded belly. Two bands of horizontal grooves at base of neck and shoulder, either size of x 2 horizontally-perforated lugs. Diam. 140 mm; 0.27 EVEs. (fill 154).

P14: Fabric GRF. Small vessel with curving profile (barrel urn or ovoid vessel; In-curved, simple rim. Undecorated. Diam. 110mm; 0.10 EVEs. (fill 163). Not illustrated.

P15: Fabric F3. Large vessel with slightly curving profile (?barrel urn) and expanded/flattened rim. Fingernail impressions to rim top and two lines of horizontal fingernail impressions below rim. 0.10 EVEs. (fill 154). Not illustrated.

P16: Fabric F4. Globular urn. Slightly everted neck and simple/rounded rim. Band of DEEP horizontal grooves at neck. 0.12 EVEs. (fill 154). Not illustrated.

P17: Fabric F4. Large vessel with slightly curving profile (barrel urn) and internally-expanded/flattened rim. Row of fingernail impressions below rim. 0.05 EVEs. (fill 154). Not illustrated.

P18: Fabric F4. Large vessel with slightly curving profile (barrel urn) and simple/squared rim. Row of fingernail impressions below rim. 0.05 EVEs. (fill 154). Not illustrated.

P19: Fabric F1. Mid-sized vessel with slightly curving profile (barrel urn) and simple/squared rim. Row of fingernail impressions below rim. 0.02 EVEs. (fill 154). Not illustrated.

P20: Fabric F4. Mid-sized vessel with curving profile

(barrel urn) and simple/squared rim. Single imperforate lenticular lug below rim. 0.07 EVEs. (fill 154). Not illustrated.

P21: Fabric F4. Base sherd from. Undecorated but with crushed flint imbedded in base underside. (fill 154). Not illustrated.

P22: Fabric GRF. Mid-sized vessel with slightly curving profile (barrel urn), upright/slightly everted neck/rim. Row of fingernail impressions at shoulder. 0.03 EVEs. (fill 154).

P23: Fabric F5. Globular urn. Upright/simple rim. Band of horizontal grooves at neck. 0.05 EVEs. (fill 154).

P24: Fabric F3. Body sherd from large vessel narrow ‘pinched-up’ horizontal strip (plain).(fill 154). Not illustrated.

Ditch 151

P1: Fabric F4. Body sherd from large vessel with fingertip-impressed applied horizontal strip. (fill 152). Not illustrated.

Ditch 174

P6: Fabric GR1. Globular urn. Upright/squared rim. Band of horizontal grooves at the neck. 2 x horizontal/lenticular (imperforate) lugs at shoulder. 0.20 EVEs. (fill 175).

Pit 158

P25: Fabric QF1. Rim form thin-walled (7 mm) vessel. Curved walls and with simple rim. Undecorated. 0.05 EVEs. (fill 160).

P26: Fabric F6. Small rim sherd form larger vessel with simple/rounded rim. Undecorated. 0.04 EVEs. (fill 160).

Roman Pottery (E.R. McSloy)

A small Roman group, comprising only fourteen sherds (85 g), was recovered in Area A. The single stratified group, from ditch 102, entirely comprised sherds in south-east Dorset Black-burnished ware, a type produced in the Poole Harbour area from the first to the fourth centuries AD. The group includes rim sherds from jars with upright or slightly everted rims, matching Seager Smith and Davies’ Types 1 or 2, and a first or second-century AD date is therefore suggested. Such a small group must limit the scope of further interpretation. The Roman-period activity identified in Area A may be peripheral to an area of occupation

228


of unknown status or extent, possibly located to the north or west of the area investigated.

Fired clay (Jacky Sommerville)

Loom weights

Two joining fragments from a horizontally-perforated cylindrical object were recovered from ditch 153, in Area D. Objects of this type, generally interpreted as loom weights, are commonly recorded from Middle Bronze Age settlement sites in southern England (Woodward 2009, 296). As conjoined, the object has an external diameter of c. 100mm, an internal diameter of c. 20 mm, and weight of 244 g. Four other conjoining fragments of similar fabric appear to represent a further loom weight.

Fragments

Five fragments of fired clay (33g) were recovered from fill 133 of pit/posthole 132, and fill 173 of ditch 178. None of these fragments retain any surfaces which might suggest an original form or function.

Metal objects (Jacky Sommerville)

Copper-alloy

Three copper-alloy objects were retrieved from the excavation. That from subsoil 109 is a rivet in good condition. The head is flat, and circular in form (diameter 21 mm), with two squared-off sides. A medieval/post-medieval date is most likely. A copper-alloy fragment from fill 166, of Middle Bronze Age ditch 178, represents the rounded tip of a strip-shaped object, which is interpreted as an awl. A small sheet fragment, recovered from topsoil 108, probably derives from a vessel of uncertain date.

The Biological Evidence

Animal Bone (Andy Clarke)

A total of 46 fragments of animal bone, (287g), was recovered by hand-excavation and bulk soil-sampling. The bone was moderately well-preserved, but highly fragmented, with much evidence of surface erosion. Some 73% of the assemblage was unidentifiable to species.

Middle Bronze Age

Five fragments (1 g) were recovered from deposit 167, the primary fill of Middle Bronze Age pit 161, none of which was identifiable to species. A further 40 fragments (283g) were recovered from secondary

fill 163. The remains of cattle (Bos taurus) and sheep/goat (Ovis aries/Capra hircus), were present in this assemblage, which was broadly typical of this period (Baker and Worley, 2014).

Plant Macrofossils and Charcoal (Sarah Cobain)

Introduction

Nine bulk soil samples were recovered from Bronze Age pits, and ditches 177 and 178 of the rectilinear enclosure. Large quantities of modern roots were obtained from the flots of six of the nine samples obtained, thus presenting a high risk of contamination through bioturbation within the samples, and therefore no radiocarbon dating was undertaken.

Results and Discussion

Middle Bronze Age: Carbonised plant macrofossils were recorded only in small quantities, with a single, poorly-preserved indeterminate grain fragment recovered from ditch slot 156, in enclosure ditch 177. In addition, a small number of poorly-preserved emmer/spelt wheat (Triticum dicoccum/Triticum spelta), possible barley (Hordeum vulgare), and indeterminate, cereal grains were recovered from the fill of storage pit 161. Charcoal was present in small, moderately well-preserved quantities within pit/posthole 132, rectilinear enclosure ditches 177 and 8 (slots 151 and 156), and pit 158. Identifications included oak (Quercus), maple (Acer campestre), alder/hazel (Alnus glutinosa/Corylus avellana), ash (Fraxinus excelsior), hawthorn/rowan/crab apple (Crataegus monogyna/Sorbus/Malus sylvestris), cherry species (Prunus) and blackthorn (Prunus spinosa). Charcoal within storage pit 161 was more abundant, but poorly preserved, and was identified as oak and hawthorn/rowan/crab apple.

The paucity and/or poor preservation of the carbonised remains precluded any interpretation of economic activities undertaken on this site. It is not possible to determine whether the recovered grain represents burnt residue within a fire-sterilised storage pit or, conversely, the reuse of this feature for disposal of domestic or crop processing waste. The identifications of limited charcoal samples indicated the sourcing of fuel from stands of mature woodland, including oak, ash and maple, with smaller amounts obtained from areas of scrub woodland, including alder/hazel, hawthorn/rowan/crab apple, blackthorn and cherry species.

229


Discussion

The 2015 excavation confirmed and qualified the results of previous field evaluations, in identifying a Middle Bronze Age ditched enclosure. The moderate assemblage of finds suggests a primarily domestic character, and a date-range of c. 1600 – 1000 BC.

A limited and poorly-preserved sample of biological material tentatively suggests a mixed farming economy, possibly involving seasonal grazing of the lower valley slopes and floodplain of the River Stour, combined with cultivation on higher contours. Excavation provided no evidence of field divisions or cultivation, although there is otherwise widespread evidence of prehistoric field systems and settlement within surrounding parts of the Stour Valley (Yates 2007, 60).

The Stour valley would have offered an attractive prospect for settlement from the Mesolithic period onwards, although, by the Middle Bronze Age, cultural considerations, including ancestry and identity, may also have contributed to patterns of land tenure (Field 2001, 58-60). The Early Bronze Age Beaker sherds, recorded from F211 of Trench 2 of the 2012 evaluation (AC 2012, 3-4), appear to represent longer-term, probably intermittent, patterns of Bronze Age activity on this site. Considerations of long-term ‘ownership’ or land tenure are problematic, as the chronology of Middle Bronze Age settlement here appears to have been relatively limited, and control of resources at this stage of prehistory may have remained relatively fluid (Brϋck 1999, 60-61;Brϋck and Goodman 1999, 7-8 ).

The relatively well-preserved Deverel-Rimbury pottery group is of moderate size, and represents a significant addition to the regional ceramic record for this period. The range of fabric types is consistent with other regional groups of this date, including the large assemblage from Bestwall Quarry, Wareham (Woodward 2009a). A further comparable regional group is represented by that from Thorney Down, Wiltshire (Ellison 1987, 385-92; Stone 1941).

The broad distinction between flint and grog-tempered fabrics from this site is also characteristic of Deverel-Rimbury groups, and may indicate locally-sourced clays (Cleal 1995, 185-94). In this case, there is a notable correlation between fabric type and vessel form, and from their contextual associations it is reasonable to assume that the two principal fabric types are broadly contemporary, although they appear to represent different clay sources, and distinct potting traditions. On this basis, the simpler grog-tempered bowl and jar forms suggest production and

distribution at a local level, whereas the finer-tempered globular urn vessels will have circulated through wider exchange networks. Contemporary regional assemblages have provided significant insights into the scale and function of exchange networks during the Middle Bronze Age period (Sharples 2010, 99; Ellison 1981, 413-38), with detailed regional models of Deverel-Rimbury societies provided, inter alia, by Ellison (1980, 127-40; 1981, 413-38), and Rowlands (1976). Both authors have emphasised the hierarchy of exchange relationships evident in the distributions of both pottery and metalwork. On this basis, the distribution-ranges of ‘everyday’ grog-tempered forms appear to be restricted to localised socio-economic units, of approximately 10-20 km in extent. By contrast, the ‘fine ware’ types are characterised by wider, regional patterns of distribution, which correlate closely with those of some classes of metalwork (Rowlands 1976, 163). Ellison has identified nine distinct types of ‘everyday’ wares in Wessex, of which the widest degree of variation was evident in those types from Dorset. Of the Deverel-Rimbury ‘fine ware’ types, a distinct regional group has been identified within the area of south Wiltshire and north Dorset embracing the Avon and Stour valleys (Ellison’s Type III, 1980, 132-38), and this might offer further scope for comparison with the material from this site.

The suggested contextual association of this material with ‘structured deposition’ (McSloy, this report) has been widely observed elsewhere for assemblages of this period (Woodward 2009, 264; Hamilton 2002). Significantly, the fills of pit 161 also contained the greater proportion of an admittedly limited animal bone assemblage, and suggest an association with feasting. The locations of both deposits reflect a frequently-observed distinction between interior and exterior aspects of enclosed domestic space, and an emphasis on house and enclosure thresholds (Wait 1985, 155; Bowden and McOmish 1987, 76-84). Hill (1995, 120), has highlighted the significance of well-preserved groups of animal bone and pottery sherds in such contexts, and has suggested that these may represent relatively unusual special events such as feasts, or possibly ‘closure events’ following abandonment.

A small, single stratified group of Roman pottery from ditch 102 (Area A) entirely comprised sherds of Dorset Black-burnished ware of probable first to second century AD date. A few additional Roman sherds, and a fragment of Roman tile, were recovered as unstratified finds. This material indicates

230


unspecified activity of this date within the vicinity, although there are otherwise no indications of Roman occupation. In view of the relatively attractive location for domestic settlement offered by fertile, well-drained valley slopes, a persistent, but discontinuous, pattern of occupation and land use seems likely to have characterised this site over a long period.

ACKNOWLEDGEMENTS

Particular thanks are due to Bellway Homes Ltd, and to the assistance of Toby Peter-Simmonds. The advice and assistance of Steve Wallis, Senior Archaeologist, Dorset County Council, is also gratefully acknowledged. The fieldwork was managed by Richard Greatorex, and was undertaken by Joe Whelan, assisted by Steve Bush and Tom Hackett. The post-excavation work was managed by Karen Walker. The illustrations were prepared by Lucy Martin and Sam O’Leary.

BIBLIOGRAPHY

AC Archaeology, 2008. ‘Results of an archaeological evaluation at Milldown Middle School, Blandford Forum’, Dorset, Unpublished AC Archaeology client report.

AC archaeology, 2012. ‘Proposed Housing Development at Former Milldown School, Blandford, Dorset: Results of an Archaeological Evaluation’, AC Typescript report.

Baker, P. and Worley, F. 2014. Animal bones and archaeology: Guidelines for best practice, English Heritage, Swindon.tfju

Bowden, M. and McOmish, D. 1987. ‘The Required Barrier’ Scottish Archaeological Review 4, 76-84.

Brϋck, J. (ed.) 2001. Bronze Age Landscapes: transition and transformation, Oxbow Books, Oxford.

Brϋck, J. and Goodman, M. 1999. Making Places in the Prehistoric World: themes in settlement archaeology, Routledge, London.

CA (Cotswold Archaeology) 2014. ‘Former Milldown School, Blandford Forum, Dorset: Archaeological Evaluation’, CA Typescript report.

CA (Cotswold Archaeology) 2015. ‘Former Milldown School, Blandford Forum, Dorset: Archaeological Excavation’, CA Typescript Report.

Calkin, B. 1964. ‘The Bournemouth Area in the Middle and Late Bronze Age, with the Deverel- Rimbury problem reconsidered’, Archaeological Journal 119, 1–65.

Cleal, R. M. 1995 ‘Pottery fabrics in Wessex in the fourth to second millennia BC’. In Kinnes and Varndell (eds.), 185-194.

Ellison, A. 1980 ‘Settlements and regional exchange: a case study’ In J. Barrett and R. Bradley (eds.), 127-40.

——— . 1981. ‘Towards a socio-economic model for the Middle Bronze Age in southern England’, in I. Hodder, G.L. Isaacs and N. Hammond (eds.) Patterns of the Past: Studies in Honours of David Clarke, 413-38, Cambridge University Press, Cambridge.

——— . 1987. ‘The Bronze Age Settlement at Thorny Down: pots, postholes and patterning’, Proceedings of the Prehistoric Society 53, 385-92.

Field, D. 2001. ‘Place and memory in Bronze Age Wessex’ In J. Brϋck (ed.) Bronze Age Landscapes: transition and transformation Oxford, Oxbow Books, 57-64.

Ford, S., Bradley, R., Hawkes, J. and Fisher, P. 1984. ‘Flint-working in the Metal Age’, Oxford Journal of Archaeology 3(2), 157–73.

Gibson, A. and Woods, A., 1997. Prehistoric Pottery for the Archaeologist, Leicester, Leicester University Press (Second Ed.).

Hamilton, S. 2002. ‘Between Ritual and Routine: interpreting British Prehistoric Pottery Production and Distribution’ in Woodward and Hill, 2002, 38-53.

Hill, J.D. 1995. Ritual and Rubbish in the Iron Age of Wessex: A Study in the Formation of a Specific Archaeological Record. Oxford, British Archaeological Reports Br.Ser. 242.

Kinnes, I. and Varndell, G. (eds.) 1995. ‘Unbaked Urns of Rudely Shape’ Essays on British and Irish Pottery for Ian Longworth. Oxford, Oxbow Monograph 55.

Ladle, L. and Woodward, A. 2009, Excavations at Bestwall Quarry, Wareham 1992–2005, Volume 1: The Prehistoric Landscape, Dorset Natural History and Archaeological Society Mono. 19, Dorchester.

Morris, E. forthcoming. ‘The Late Bronze Age Pottery’ in Cotswold Archaeology, forthcoming, Excavations at Hartshill Copse, Upper Bucklebury, West Berkshire.

Rowlands, R. 1976. The Production and Distribution of Metalwork in the Middle Bronze Age in Southern Britain. Oxford, British Archaeological Reports Br.Ser. 31.

Seager Smith, R. and Davies, S. M. 1993, ‘Roman Pottery’, in Woodward et al. 1993, 202–14.

Sharples, N., 2010. Social Relations in Later Prehistory. Wessex in the First Millennium BC. Oxford, Oxford University Press.

Stone, J.F.S, 1941. ‘The Deverel-Rimbury settlement on Thorny Down, Winterbourne Gunner, S. Wilts’, Proc. Prehist. Soc. 4, 114-33.

Wait, G., 1985. Ritual and Religion in Iron Age Britain. Oxford, British Archaeological Reports Br. Ser. 149.

Whittaker, J. C. 1994. Flintknapping: Making & Understanding Stone Tools. Austin. University of Texas Press.

Woodward, A. 1995. ‘Vessel size and social identity in the Bronze Age of southern Britain’. In Kinnes and Varndell (eds.), 195-202.

——— . 2009a. ‘The Pottery’, in Ladle and Woodward 2009, 200–271.

——— . 2009b. ‘Fired clay’, in Ladle and Woodward 2009, 289–301.

Woodward, A. and Hill, J.D. (eds) 2002. Prehistoric Britain: the ceramic basis Oxford, Oxbow Books.

Woodward, P.J., Davies, S.M. and Graham, A.H. 1993. Excavations at Greyhound Yard, Dorchester 1981–4. Dorset Natural History and Archaeological Society, Dorchester.

Yates, D.T. 2007. Land, Power and Prestige. Bronze Age field systems in southern England. Oxford, Oxbow Books.

231


excavations at former milldown school ......introduction in march and april 2015, cotswold...

Documents